PROSPECTIVE-STUDY OF SERUM MICRONUTRIENTS AND OVARIAN-CANCER

Background: Antioxidant micronutrients, such as alpha-tocopherol (vita min E), the carotenoids, and selenium, may protect against the develop ment of cancer by preventing free radical damage at the cellular level . Purpose: A nested case-control study was conducted among donors to a serum bank to examine the association between levels of serum micronu trients and/or cholesterol and the development of ovarian cancer. Meth ods: In 1974, sera were collected from 20 305 residents of Washington County, MD, over a 4-month period and stored at -70 degrees C, Serum m icronutrient concentrations of women who developed ovarian cancer (cas e subjects, n = 35) were compared with those of women who remained fre e of cancer and who were matched to case subjects on age and menopausa l status (control subjects, n = 67). Serum levels of retinol (vitamin A), alpha- and beta-carotene (the major provitamin A), lycopene (a car otenoid), and alpha- and gamma-tocopherol were measured using high-per formance liquid chromatography. Serum selenium was measured by neutron activation analysis. Cholesterol was measured by enzymatic assay. The data were categorized into thirds and conditional logistic regression analyses were performed to determine the association between prediagn ostic serum cholesterol and micronutrient levels and the development o f ovarian cancer; matched odds ratios (ORs) were determined from these regression analyses. P values for trend and for interaction were calc ulated with the use of two-sided likelihood ratio tests. Results: High er serum alpha-tocopherol levels were associated with an increased ris k of ovarian cancer (P for trend = .04); however, this association dim inished after adjustment for cholesterol. Women with higher serum chol esterol levels had an increased risk of ovarian cancer compared with w omen in the lowest third of cholesterol levels (OR = 3.2; 95% confiden ce interval = 0.9-11.3). The association between serum cholesterol lev els and the risk of ovarian cancer was examined, stratifying by micron utrient level. The general pattern observed was an increased risk of o varian cancer associated with cholesterol levels greater than 200 mg/d L, regardless of the micronutrient level. Serum selenium was associate d with a decreased risk of ovarian cancer only among case participants diagnosed 4 or more years after blood collections (P for trend = .02) . Concentrations of carotenoids and retinol were not associated with t he development of ovarian cancer. Conclusions: Selenium may have a pro tective role against the development of ovarian cancer. Higher serum c holesterol levels were associated with an increased risk of developing ovarian cancer; an association that persisted regardless of serum mic ronutrient level. Implications: Given the small size of this study and the inconsistency of results among the few prospective studies of ova rian cancer conducted to test these associations, replications of thes e Endings are highly desirable.